An17 exhibitor insights_tuesday_whats hiding in your point of sale data

19
RETAIL’S BIG SHOW 2017 | #NRF17 What’s Hiding in Your Point of Sale Data? IRAD BEN-GAL, Professor and Chairman, Stanford University/C-B4 JOE GAUTHIER, Director, Operations, Wesco, Inc. MIKI CISIC, Director, Sales, C-B4 Analytics

Transcript of An17 exhibitor insights_tuesday_whats hiding in your point of sale data

Page 1: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17 What’s Hiding in Your Point of

Sale Data?

IRAD BEN-GAL, Professor and Chairman, Stanford University/C-B4JOE GAUTHIER, Director, Operations, Wesco, Inc.

MIKI CISIC, Director, Sales, C-B4 Analytics

Page 2: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

The content of this document is C-B4 Confidential & Proprietary. If you are not the intended recipient and have received this document, any use or distribution is prohibited. Please notify [email protected] immediately by e-mail and delete this message from your computer system.

What's Hiding in Your Point of Sale Data?Explore the differences between market basket analysis vsin-store consumer behavior

Page 3: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

Basket Analysis (a.k.a Affinity Analysis)• Which group of items are likely (or less

likely) to be purchased together? For example, beer & potato-chips or shampoo & conditioner….

• Provides a better understanding of the individual purchase behavior of the customer (“impulsive customer purchase”).

• Well established & helpful in many applications

P R E D I C T I O N S & I N S I G H T S T H R O U G H P A T T E R N S3 |

C C

Page 4: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

Market Basket Analysis vs. In-store Purchase Pattern Analysis

P R E D I C T I O N S & I N S I G H T S T H R O U G H P A T T E R N S4 |

Page 5: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

BreadButter

Milk DiapersBeer{Bread & Butter Milk}

Mon #1

Mon #2

Tue #1

Tue #2

Wed #1

Wed #2

Thu #1

Thu #2

• Probability {Milk}: 5/8 ~ 60%• Probability {Milk given Bread & Butter }: 2/2

= 100%

Can we learn more at an aggregated daily/store level?

{Beer Diapers} • Probability {Diapers}= 6/8 =75% • Probability {Diapers given Beer)= 3/4=75%,

Lift = 0%

• Lift=40%: (When selling Bread & Butter the probability for selling milk increases by 40%, but it applies to only 2 transactions)

P R E D I C T I O N S & I N S I G H T S T H R O U G H P A T T E R N S

Purchase Example (sourced from Wikipedia)

• No correlation is found at a transactional level

5 |

Page 6: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

P R E D I C T I O N S & I N S I G H T S T H R O U G H P A T T E R N S

Purchase Example (cont.)

• Reflects a non-transactional consumption preference pattern

{Beer Diapers} are correlated at a store level

• Is this pattern significant (statistically)?

• How can we use it at the chain level to analyze the stores’ performance?

• Let’s check this pattern across all stores

6 |

BreadButter

DiapersBeer

Mon #1

Tue #1

Wed #1

Thu #1

Milk

Mon #1

Mon #2

Tue #1

Tue #2

Wed #1

Wed #2

Thu #1

Thu #2

Page 7: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

Consumer Purchasing Patterns at a Store level(Beer & Diapers example - out of trillions of combinations)

• Automatically analyzingmillions of consumption preference patterns

• Root cause of anomalies:

- Operational failures

- Availability issues- Other local effects

P R E D I C T I O N S & I N S I G H T S T H R O U G H P A T T E R N S7 |

Page 8: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

Basket Analysis (Individual behavior)• Focused on individual purchase behavior of a customer • Personal Applications: Cross-Selling & Up-Selling, personalized coupons,

personalized emails• General Applications: Store Design, Loyalty programs, Promotional plans…

P R E D I C T I O N S & I N S I G H T S T H R O U G H P A T T E R N S8 |

Bottom Line: Benefits of both Methods

In-Store Purchase Pattern Analysis (store behavior) • Does not require transactional & personalized data (faster & simpler POS data)• Reflects non-transactional patterns (even by different customers)• Lower Error Rate (aggregation reduces false rules by random associations of

products)• More effective for store behavior analysis – correcting operational failures,

availability issues, localizing assortment, analyzing local effects

Page 9: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

DEMO

Page 10: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

Proven to increase same store growth by 0.8 to 3%

Uncover local consumer purchasing patterns within simple sales and inventory data in order to:

• Fine tune assortment at a store level to better fulfill local preferences

• Detect + correct in store operational anomalies that prevent high volume sales

10

No Hardware

Automated

No External Data

Page 11: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

• Chain of 52 convenience stores in Muskegon, Michigan• Family owned• Started by Bud Westgate in 1952 with a single store and 3 used gas pumps• Continuous growth, expanded to include:

• Distribution center• Central bakery and deli• 6 Subway locations• Bulk fuel and propane business + Wesco Energy division

• For 55 years the mission has been Q-PPAS: Quality People, Products, Associates and Service

P R E D I C T I O N S & I N S I G H T S T H R O U G H P A T T E R N S1 4 |

Page 12: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

P R E D I C T I O N S & I N S I G H T S T H R O U G H P A T T E R N S1 5 |

Pilot: Your total effort is less than a day

2 hours

Extract Data

Wesco

Generate Initial Recommendati

ons

3 hours

CB4Filter &

DistributeRecommendati

ons

1/2 Day

TogetherROI and

A/B Testing Report

5 min

CB4Recommendations

Sent to select stores.

Feedback collected

On-going

Stores

Page 13: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

1-2 DAYS• Installation• Configuration of data

extract• Integration of business

constraints into the analysis to increase the relevance of recommendations for store and merchandising managers

Deployment in less than a week

P R E D I C T I O N S & I N S I G H T S T H R O U G H P A T T E R N S1 6 |

1. Setup & Installation

2. Training 3. App Deployment

Uncovers behavioral patternsand translates them into recommendations*Available on premise as well

5 DAYS• Training

- Solution owner/users – 5h- District manager – 3h- Store managers – 0.5h- Merchandising team – 2h

• Perform store rides• Perform several dry runs• Schedule automatic deployment

of recommendations to stores via email, back office PC, or mobile app

1 HOUR• Installation of back office or

mobile app• App is straightforward and

does not require dedicated training (comes with manual and video clips that walk the store managers through the process)

Page 14: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

Web ConsoleAbility to schedule and deploy

recommendations + review dashboards that measure ROI and

revenue lift

Point of Sale Data

CB4 ServersUncovers behavioral patterns

and translates them into actionable recommendations

Store managers/supervisorsRecommendations to resolve

operational opportunities

Merchandising ManagersRecommendations that help to

localize assortments and planograms

Feedback from stakeholders is tracked and measured

P R E D I C T I O N S & I N S I G H T S T H R O U G H P A T T E R N S1 7 |

Page 15: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

App - Demo

P R E D I C T I O N S & I N S I G H T S T H R O U G H P A T T E R N S1 9 |

Review your list of “Open Tasks”

Tap on each task to see details

Submit findings 

Page 16: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

P R E D I C T I O N S & I N S I G H T S T H R O U G H P A T T E R N S2 0 |

$1,910,432Estimated annual revenue

Lift2%

Operational opportunities detected 582

Page 17: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

P R E D I C T I O N S & I N S I G H T S T H R O U G H P A T T E R N S2 1 |

HIT RATE

Page 18: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

DYNAMICALLY LOCALIZED

ASSORTMENT• 200 new items

introduced to assortment• Improved planogram

execution

IMPROVED CUSTOMER EXPERIENCE

• Store associate awareness to product display & operational opportunities.

• Weekly feedbacks

0.8 – 3% SAME STORE

GROWTH• On track to $2M

same store growth by end of fiscal year (2.2% sales increase)

IMPROVED AVAILABILITY• Increase

availability in stores by 7%

• Better Supply chain decisions

Value & Benefits

P R E D I C T I O N S & I N S I G H T S T H R O U G H P A T T E R N S2 2 |

Page 19: An17 exhibitor insights_tuesday_whats hiding in your point of sale data

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

RE

TA

IL’S

BIG

SH

OW

20

17

|

#N

RF

17

Visit us at booth #4252 for more information